Coping with Heart Failure: New Opportunities on the Horizon
September, 2003
Researchers have long been interested in the development, progression and treatment of heart failure. They have created mouse models of the disease, designed devices to assist the heart in its pumping activity, and developed drugs to boost heart function. Yet the underlying mechanisms of heart disease remain uncertain and the therapies to treat it are still imperfect.
Seeking to gain a better understanding of heart failure and find new ways to diagnose and treat it, Burns C. Blaxall, formerly at Duke University in Durham, North Carolina and now at Rochester University in New York, and his colleagues are examining the activity of genes in failing and recovering hearts.
"We are very interested in finding both the causes and effects of heart failure, and one of the ways to do that is to look at changes in gene expression," he says.
For example, in a recent study, researchers asked how gene expression in the heart changes as heart failure progresses. Blaxall and his colleagues at Duke University extracted heart tissue from mice carrying mutations that program them to develop heart failure. Using mice at both early and late stages of disease, the team searched for genes whose activity might predict how heart failure proceeds.
To examine gene activity in the mutant mice, the researchers used GeneChip Mu11K Arrays. Arrays such as these allow researchers to monitor the expression of thousands of genes simultaneously and, because of their unique design, provide more reliable results than other types of microarrays.
"Affymetrix provides very nice checks and balances, controls for the readout you obtain from the microarray," says Blaxall. "I was always very impressed by the design of the Affymetrix arrays over other arrays which have a higher probability for non-specific interactions."
Surprisingly, Blaxall and his colleagues found that the patterns of gene expression were very similar in the early and late stages of disease.
"Initially, we were very depressed because there were so few changes that were predictive of the progression," says Blaxall. But then, the team compared the results to data from previous experiments using normal, healthy mice, and realized that, in fact, their findings had potentially important clinical implications.
They discovered that, at least in these mice, the pattern of gene expression that distinguishes healthy hearts from failing hearts is set at a very early stage, well before the major symptoms of the disease develop. If the same occurs in humans, patients who are likely to develop heart failure symptoms could be identified and treated early to ameliorate the disease's progression.
"The caveat is that these are mice that are programmed to get heart failure, so whether or not we can extend this to the human condition, I'm not sure," cautions Blaxall. "But it is conceivable that some patients are programmed to get heart failure, and that's something to test."
The finding was made possible through the comparison of new data and old data, an analysis which cannot be easily done when using other types of microarrays. GeneChip arrays yield expression values that can be easily and reliably compared between studies, avoiding the need to repeat experiments each time a researcher wants to make a new comparison. "That's the beauty of Affymetrix' arrays," says Blaxall. "That's the reason it will always be my chosen format for microarray processing. It's the best way to have your data so that you can do comparisons."
Indeed, Blaxall and his co-workers have used GeneChip arrays for making many other comparisons, including ones using human heart tissue. For example, they recently compared patterns of gene activity in a small number of patients suffering from heart failure before and after they received a mechanical pump called a left ventricular assist device (LVAD). LVADs are often used as 'bridges to heart transplant' to sustain patients waiting for a heart transplant because their left ventricle, which pumps blood throughout the body, is too weak to perform its normal function. Recent studies suggest that, in some cases, LVADs may even provide long-term support, serving as 'bridges to recovery.'
Using GeneChip HuGeneFL Arrays, the researchers found that LVADs seem to induce dramatic changes in gene expression in the heart. These changes probably reflect the heart's recovery response to the 'vacation-like' period offered by LVAD support, says Blaxall. If confirmed by future studies with more patients, the findings could serve as a first step towards identifying genes that predict how individual patients will respond to LVAD support.
In addition, the researchers obtained insights beyond their study's original goals, as is often the case in microarray studies. "Not only were we able to distinguish heart failure before and after ventricular assist device, but we could also pick out two different types of heart failure: the ischemic and dilated cardiomyopathies," says Blaxall. "This unexpected finding may be particularly relevant to the clinic. It shows us that even at the level of molecular gene expression these cardiomyopathies are different types of heart failure, and it may behoove us to treat them differently." Currently, both diseases are treated very similarly.
The team is now examining individual genes highlighted by their studies in greater detail , and bringing together data from several groups to create larger databases on heart gene expression. In addition, combining their studies in humans and mice, they are searching for genes whose expression is associated with healing hearts in both species. "If we can find genes that are common, then we may have a core set of genes that may have very high diagnostic or therapeutic value," says Blaxall. With their proven track record, GeneChip arrays promise to help researchers such as Blaxall reach their ambitious goals.
(1) Blaxall, B.C., Spang, R., Rockman, H.A., and Koch, W.J. Differential myocardial gene expression in the development and rescue of murine heart failure. Physiol. Genomics (2003- article in press - electronic version available before print).
(2) Blaxall, B.C., Tschannen-Moran, B.M., Milano, C.A., and Koch, W.J. Differential gene expression and genomic patient stratification following ventricular assist device support. J Amer. Coll. Cardiol. 41(7):1096-106(2003).
The author, Marina Chicurel, interviewed Burns Blaxall in August, 2003. She received her Ph.D. in Neurobiology at Harvard University and is a contributing writer for the Hereditary Disease Foundation, Nature, and Science.
